[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124287-en":3,"doc-seo-124287-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124287,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","MACHINE LEARNING MODEL FOR CORRELATING MICROSTRUCTURAL FEATURES AND MACROSCOPIC PROPERTIES OF HETEROGENEOUS COMPOSITES","Machine learning models are developed to predict macroscopic properties of heterogeneous composites where conventional micromechanics parameters cannot fully capture complex microstructure–property relationships. Two approaches are used: feature-engineered ANN models based on SEM-derived microstructural descriptors for hardness prediction in nickel-based superalloys, and non-feature-engineered CNN models that learn directly from microstructure images to predict effective thermal conductivity of thermal insulation composites. The study shows improved accuracy over physics-based models and reliable generalization to novel microstructures.","MACHINE LEARNING MODEL FOR CORRELATING MICROSTRUCTURAL FEATURES AND MACROSCOPIC PROPERTIES OF HETEROGENEOUS COMPOSITES  \nCHENGCHENG SHEN1,2, HAIFENG ZHAO1,2,†, RUINAN MU1,2, ANPING WANG1,2,  \nKE WANG1,2  \n1University of Chinese Academy of Sciences  \n2Key Laboratory of Space Utilization, Technology and Engineering Center for Space Utilization,  \nChinese Academy of Sciences  \nBeijing, China  \n† [Corresponding email: hfzhao@csu.ac.cn](Corresponding email: hfzhao@csu.ac.cn)  \nKey words: Machine learning, artificial neural network (ANN), convolutional neural network (CNN), nickel base superalloys, fibrous and particulate composites, thermal conductivity, hardness.  \nSummary. This study explores the use of machine learning (ML) models in predicting the macroscopic properties of heterogeneous composites. Traditional micromechanics parameters have limitations, thus ML models with and without feature engineering are utilized. For artificial neural network (ANN) models with feature engineering, microstructural descriptors from SEM images of nickel-based superalloys are used to predict hardness. 10 descriptors are selected to reduce the computational cost of the deep neural network (DNN) with the support of the shallow neural network (SNN), and accuracy is enhanced by incorporating two additional descriptors. The result surpasses existing physics-based models. Models without feature engineering employ a convolutional neural network (CNN) to predict the effective thermal conductivity of thermal insulation composite materials. The CNN model demonstrates accurate predictions for novel microstructures. ML models can achieve more efficient predictions than traditional methods, indicating their potential in advancing materials science. In summary, harnessing artificial intelligence to capture the scattering characteristics of heterogeneous materials enables both DNN and CNN models to achieve more efficient predictions compared to traditional methods. This highlights the potential of machine learning in advancing materials science and expediting the development of materials with desired properties.  \n1 INTRODUCTION  \nComposite materials with complex multi-scale microstructure typically exhibit great performance[1, 2] . At the same time, the complex multi-scale microstructure also hinders the accurate description and performance prediction of computational modeling of composite materials [3, 4] . Traditional computational models are constrained by limited input parameters, restricting their capacity to extract features from complex microstructures and effectively describe heterogeneous composites. [5] . For example, in the traditional micromechanics context,  \nthe size and volume fraction of inclusions have long been regarded as the primary microstructural parameters dictating macroscopic properties[6-8] . Obviously, traditional models frequently struggle to achieve a comprehensive understanding of the intrinsic characteristics of materials, let alone analyze the impact of each feature on macroscopic properties.  \nMachine learning (ML) automates the creation of computing models using training data, rather than relying on manually coded programs with predetermined logic. [9] . Therefore, with the accumulation of material experimental data and microstructural characterization, ML methods can be used to accurately and quickly perform multi-scale calculations of materials and correlate the material microstructure and macroscopic properties[10] . It also can infer the composition[11], microstructure[12], and even preparation process parameters[13], etc. from the macroscopic properties of the material, which can provide guidance for reverse design of materials[14, 15] . The integration of ML techniques into materials research is a key focus in modern scientific investigation, inspired by AI4Science [16, 17] .  \nArtificial neural network (ANN) is one of the commonly used machine learning models, which can establish a complex nonlinear mapping between i","cbCaiprhw0hTtaIp","https://ap.wps.com/l/cbCaiprhw0hTtaIp","pdf",1204147,1,12,"English","en",105,"# Introduction\n## Motivation and limitations of traditional micromechanics\n## Machine learning for microstructure–property correlation\n## ANN vs. CNN and feature engineering\n## Scope of this study and overall workflow","[{\"question\":\"Why are traditional micromechanics-based models limited for predicting properties of heterogeneous composites?\",\"answer\":\"They rely on a restricted set of input parameters and struggle to extract comprehensive features from complex multi-scale microstructures, limiting insight into how individual microstructural characteristics affect macroscopic properties.\"},{\"question\":\"How does the feature-engineered ANN model predict hardness?\",\"answer\":\"It uses selected microstructural descriptors extracted from SEM images of nickel-based superalloys as inputs to an ANN, with descriptor selection and augmentation to improve accuracy and computational efficiency.\"},{\"question\":\"How does the CNN model predict effective thermal conductivity without feature engineering?\",\"answer\":\"The CNN takes microstructure images directly as input, automatically learning image features that correlate with effective thermal conductivity, and it can produce accurate predictions for novel microstructures.\"}]","MACHINE LEARNING MODEL FOR CORRELATING MICROSTRUCTURAL FEATURES AND MACROSCOPIC PROPERTIES OF HETEROGENEOUS COMPOSITES | PDF",1785821380,30,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-model-for-correlating-microstructural-features-and-macroscopic-properties-of-heterogeneous-composites","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-model-for-correlating-microstructural-features-and-macroscopic-properties-of-heterogeneous-composites/124287/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are traditional micromechanics-based models limited for predicting properties of heterogeneous composites?","Question",{"text":75,"@type":76},"They rely on a restricted set of input parameters and struggle to extract comprehensive features from complex multi-scale microstructures, limiting insight into how individual microstructural characteristics affect macroscopic properties.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the feature-engineered ANN model predict hardness?",{"text":80,"@type":76},"It uses selected microstructural descriptors extracted from SEM images of nickel-based superalloys as inputs to an ANN, with descriptor selection and augmentation to improve accuracy and computational efficiency.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the CNN model predict effective thermal conductivity without feature engineering?",{"text":84,"@type":76},"The CNN takes microstructure images directly as input, automatically learning image features that correlate with effective thermal conductivity, and it can produce accurate predictions for novel microstructures.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":29,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]